Computer Science (arXiv)

Custom AI agents areagents that live inside their own application, talk to their own data and tools, enforce their own security boundaries, and carry their own brand and audit trail. What separates them from the general-purpose tier is fit, not capability: each is built for one job, by the engineer who will maintain it. No published practice sets out how to build one end to end. The pieces are everywhere (function-calling APIs, the Model Context Protocol, code agents to pair with), but the practice that chains them lives in podcasts, blogs, and leaked system prompts. This paper writes that practice down as a methodology, Agents All the Way Down: two preconditions crossed once and kept, then three practices repeated for the agent's life. The preconditions are (P1) Substrate, the LLM as a software component, framed as tools, then system, then messages under prompt-caching; and (P2) Building blocks: function calling, MCP, CLI orchestration, the liteshell pattern, the agent loop, skills, characters, hooks, and scaffolding. The practices are (P3) prototype with a general-purpose agent; (P4) harvest, fold, and ship the result as a CLI, the Turtle pattern; and (P5) agent-tests-agent, in which a general-purpose agent drives it through behavioural scenarios, a complement to classical testing, not a replacement. The working loop is P3 to P4 to P5 and back, and one corollary falls out for free: multi-agent orchestration is just CLI composition. The methodology is framework-free by construction. It was distilled from the AAC, a custom agent for the open-source LAMB platform, built in about ten days by one developer with an AI pair-programmer and in production . We present it as a transferable practice, independent of any language or framework.

De novo peptide sequencing from tandem mass spectrometry is pivotal in proteomics, enabling identification of novel peptides without reference databases. While recent Transformer-based encoder-decoder models have achieved remarkable performance, we uncover a critical pathology in their inference dynamics. Through comprehensive feature scaling experiments, we demonstrate that existing auto-regressive peptide decoders tend to over-rely on generated-sequence priors while progressively under-utilizing fine-grained physical evidence from the input mass spectrum. This phenomenon leads to suboptimal results, where generated peptide sequences are biologically plausible yet not faithful to the input spectrum. To rectify this, we propose MemNovo, a training-free and plug-and-play mechanism that re-balances peptide and spectral contributions at inference time. MemNovo alleviates the information bottleneck by establishing a persistent spectral memory bank and injecting retrieved features directly into the final decoding stage via an ultra-conservative residual connection. Theoretical analysis confirms that this mechanism restores the mutual information between the decoder state and the raw spectrum. Extensive experiments on the Nine Species benchmark with two representative baselines, Casanovo and InstaNovo, demonstrate that MemNovo consistently improves both amino acid precision and peptide precision, achieving up to 39.1% relative improvement in peptide precision for Casanovo and up to 3.9% for InstaNovo, with negligible computational overhead.

Mixture-of-Experts (MoE) and reinforcement learning (RL) post-training now dominate large language model (LLM) development, yet expert load imbalance remains a critical challenge. Existing load-balancing systems target pre-training by relying on historical step-level statistics. However, these methods fail under the unique workload dynamics of RL post-training: the step-level load is stable, but the tiny batch sizes processed during micro-steps cause severe, high-frequency load fluctuations. We introduce ForeMoE, a micro-step-level load balancing system for MoE RL post-training. Instead of relying on historical statistics, ForeMoE exploits the multi-stage RL pipeline (rollout, recompute, policy update) by using foreseeable routing information from the rollout stage to proactively guide load balancing in the remaining stages. To support frequent per-micro-step reconfiguration, ForeMoE employs a hierarchical planner that decomposes the NP-hard load balancing problem into tractable sub-components, alongside a transfer engine that leverages complementary hardware paths (CPU-assisted and GPU-direct) for overlapped expert transfer. Evaluations on 64 GPUs demonstrate that ForeMoE achieves up to a 1.45$\times$ speedup over state-of-the-art RL post-training systems.

Conformal Bayes combines Bayesian posterior predictives with conformal calibration to produce prediction sets that are both statistically valid and geometrically efficient. We study conformal Bayes under label shift from a unified perspective, identifying two complementary approaches that restore nominal target-domain coverage through importance-weighted conformal calibration but operate through independent mechanisms. \emph{Post-hoc calibration} tilts the posterior predictive toward the target domain and corrects the conformal threshold via an importance-weighted quantile, leaving the parameter posterior unchanged. \emph{In-training adaptation} tilts the parameter posterior itself to the target domain, producing a corrected predictive whose highest predictive density region serves as the highest predictive density (HPD) based prediction set under the fitted target predictive; efficiency is model-dependent and does not imply finite-sample conditional optimality. Two controlled experiments show that in an unbiased training regime both strategies achieve valid coverage equally, while in a lead-optimization regime in-training adaptation acts as a debiasing operator, reducing interval width at unchanged coverage.

Code retrieval is becoming central to coding agents, but agentic coding requires more than matching a natural-language query to an isolated snippet. Given a user request, a coding agent needs to navigate a concrete repository state, locate relevant files and functions, gather supporting context, and filter similar in-repository distractors. Existing code retrieval benchmarks mainly evaluate docstring-to-function or snippet-level matching, thereby missing this requirement-driven repository search problem. To address this gap, we introduce CORE-Bench, a comprehensive benchmark for code retrieval in the era of agentic coding. CORE-Bench evaluates code retrieval ability at three levels: code understanding, issue-to-edit localization, and broader context retrieval. Built from curated code-search tasks and SWE-bench-series instances, CORE-Bench contains over 180K queries and 106K broader-context relevance labels. Experiments with representative embedding models show a sharp drop from traditional code search to code retrieval in agentic coding settings. Simple supervised fine-tuning of existing embedding models significantly improves performance in this setting, suggesting substantial room for further progress.

Large language models (LLMs) have shown great promise for automated code translation, yet existing approaches often rely on token-level statistical patterns rather than sufficient understanding of program semantics. As a result, translated programs may still contain logical and semantic errors. Although high-quality semantic guidance, such as functional descriptions and test cases, can help mitigate these errors, such resources are often unavailable in real-world scenarios. This raises two key challenges: how to construct rich semantic information directly from source code, and how to ensure that such semantics are accurate and reliable enough to guide translation.To address these challenges, we propose Multisage, a multi-semantic augmentation and self-calibration framework for LLM-based code translation. Multisage consists of three modules. First, a semantic representation parsing module extracts structured base semantics from source code, including data-flow graphs, type constraints, and external API information. Second, a multi-semantic augmentation module builds on these representations to generate diverse augmented semantics, including code summaries, function-level test cases, and API-oriented descriptions and tests. Third, a semantic consistency calibration module uses semantics-preserving mutations and cross-semantic consistency verification to filter, calibrate, and refine the generated semantics.Experiments on the HumanEval-X code translation benchmark show that Multisage improves translation success rates by up to 2.22 times across diverse backbone models. It consistently outperforms vanilla prompting, instruction-tuned LLMs, and Chain-of-Thought reasoning, with the largest gains observed on smaller models. These results demonstrate that explicit semantic augmentation can substantially improve the reliability of LLM-based code translation.

In this paper, we introduce Representation Prediction via Autoencoding using Iterative Refinement (RePAIR) - a novel self-supervised representation learning architecture that synthesizes Masked Autoencoders (MAE), Joint Embedding Predictive Architectures (JEPA), and Bidirectional Encoder Representations from Transformers (BERT). We demonstrate how it can be used to encode objects in sequential data like consecutive chess positions into compact yet meaningful representations. The basic principle of the architecture is to mask large portions of a sequence of latent states, similar to BERT and MAE. Then, we apply a lightweight Predictor to the latent representations that repairs gaps in the sequence in a lower-dimensional embedding space akin to JEPA. Our experiments in the domain of chess show that the Encoder refines the board representations such that meaningful chess concepts emerge clustered in the latent space. Furthermore, reconstructions of the masked board states show that the model is able to reason about the piece movements without relying on costly reinforcement learning methods. Lastly, we find that the resulting representation space allows for quick and intuitive dissections of chess games by observing the game path trajectories in this semantically rich space.

Multi-channel mixed-SNR training improves out-of-distribution (OOD) generalisation of deep learning channel estimators for IEEE 802.11p vehicular communications, yet the internal mechanism responsible for this remains unexplained. This work presents REACH (Relevance-based Explanation and Architectural Compression for cHannel estimators), a gradient-based interpretability framework that operates at two levels. Input-level attribution identifies a subset of time-frequency features consistently relevant across all evaluated channel conditions, enabling input dimensionality reduction with minimal performance loss. Filter-level attribution reveals a near-universal internal representation, providing a representational account of the observed OOD generalisation. Guided by the resulting filter taxonomy, relevance-guided architecture compression substantially reduces both the number of parameters and the number of floating-point operations (FLOPs) with sub-1 dB normalised mean square error (NMSE) degradation, and OOD generalisation degrades more slowly than within-distribution accuracy under increasing compression.

There are two main Parameter-Efficient Fine-Tuning (PEFT) techniques for Large Language Models (LLMs). While Low-Rank Adaptation (LoRA) introduces additional weights between the LLM layers, Soft Prompting introduces additional fine-tuning-specific raw tokens to an LLM input. However, both require modification to the computational graphs of precompiled, preoptimized LLMs. As a result, neither is fully supported in high-throughput engines like vLLM. We propose fine-tuning with ART (Art-based Reinforcement Training). The method injects information into a frozen Multimodal Large Language Model (MLLM) by optimizing only its raw visual input, thus enabling the soft-token approach on pre-compiled computational graphs. It relies on backpropagation of gradients back into a plain pixel array and thus supports any fine-tuning objective. Moreover, the optimized visual input can be stylized as task-relevant computational artworks. The approach's effectiveness is confirmed for different sizes of a popular open Qwen architecture and for several textual benchmarks. Specifically, ART reaches accuracy competitive with LoRA across mathematics and structured-tool-use benchmarks.

Learning compatible representations aims to learn feature representations that can be used interchangeably over time whenever a model undergoes updates. In this paper, we demonstrate that stationary representations learned by d-Simplex fixed classifiers imply compatibility as in its formal definition. This result establishes a foundation for future works and can be directly exploited in practical learning scenarios. We address the challenge of learning compatibility using $d$-Simplex fixed classifiers when the model is sequentially fine-tuned. Learning according to a d-Simplex fixed classifier with the cross-entropy loss aligns feature distributions at the first-order statistics. Consequently, it may not fully capture higher-order dependencies in the representation between model updates. To address this issue, we demonstrate that training the model using a $d$-Simplex fixed classifier through a convex combination of the cross-entropy loss and a contrastive loss not only captures higher-order dependencies, but is also equivalent to learning with the cross-entropy under the compatibility constraints. We confirm our findings with extensive experiments also considering a new scenario where a pre-trained model is sequentially fine-tuned and occasionally replaced with an improved model. We show that stationary representations enable uninterrupted retrieval services (without reprocessing gallery images) while improving performance during model updates and replacements, achieving state-of-the-art. Code at https://github.com/miccunifi/iamcl2r.

Multi-modal large language models (MLLMs) depend on in-context learning (ICL) for rapid task adaptation, but their scalability is severely limited by finite context windows and the growing cost of key-value (KV) caches in long multi-modal sequences. Existing memory compression approaches typically rely on rigid token removal or sample-dependent importance estimation, which introduces bias, disrupts semantic structure, particularly for visual representations, and yields static memories that cannot adapt to new queries. We introduce TASM (Task-Aware Structured Memory), a training-free framework that addresses these limitations through task-aware, structure-preserving, and dynamically accessible memory construction. TASM employs task-vector guided compression to replace sample-specific signals with a task-level direction that captures shared relevance across demonstrations. To preserve the underlying manifold, it applies semantics-aware token merging via bipartite graph matching, aggregating tokens without destructive pruning. Finally, TASM structures memory into a hierarchy comprising a compact Core Memory and a Latent Bank, facilitating query-adaptive dynamic retrieval. Evaluations confirm TASM maintains high performance under heavy compression, effectively balancing efficiency with adaptability.

For a set $X$ of integer points, the relaxation complexity $\operatorname{rc}(X)$ is the smallest number of facets of any polyhedron $P$ such that $P \cap \mathbb{Z}^d = X$. In this paper, we focus on the case where $X$ is the discrete standard simplex $Δ_d = \{\mathbf{0}, \mathbf{e}_1, \dots, \mathbf{e}_d\}$. We show that $\operatorname{rc}(Δ_d) = O(\log d)$ by an explicit, elementary construction. This improves upon the previously best-known upper bound $\operatorname{rc}(Δ_d) = O(d / \sqrt{\log d})$ due to Aprile, Averkov, Di Summa, and Hojny (2024) and matches an asymptotic lower bound by Averkov and Schymura (2022).

Open-ended scientific discovery asks agents to move beyond executing analyses for predefined questions. Across multiple rounds of exploration, a discovery agent must decide which phenomena warrant investigation while avoiding overinterpretation, where emerging claims exceed the evidential scope of the analyses supporting them. This creates an evidence-calibration problem: the exploration trajectory must be coupled with claim status so that evidence can guide both what to investigate next and what can be claimed. We introduce StatefulDiscovery, a discovery framework that externalizes investigation state and uses it to coordinate frontier selection, evidence acquisition, and claim adjudication. We evaluate StatefulDiscovery across 40 real-data discovery tasks. Compared with several baselines, StatefulDiscovery produces more claims overall judged to be both well-supported and high-value. Ablations indicate that structured hypotheses, local adjudication, and frontier control contribute to performance. Together, these results suggest that explicit discovery state can couple exploration with evidence-calibrated claim formation.

Post-training quantization (PTQ) is essential for efficient large language model inference, but reliably quantizing activations remains challenging when weights, activations, and KV caches are all quantized to 4-bit precision. A key difficulty lies in massive activations, whose extreme values dominate the activation range and amplify quantization errors. State-of-the-art methods mainly mitigate massive activations through transformation-based smoothing, such as orthogonal rotations and affine scaling, but overlook the cross-layer dynamics of the residual stream. In this paper, we show that massive activations emerge and disappear in a phase-wise pattern across network depth, triggering large residual changes. These changes cause newly injected layer-wise updates to dominate the 4-bit quantization scale and weaken historical residual information. To characterize this behavior, we introduce Jump Ratio and Historical Feature SNR. This suggests that static transformation-based smoothing cannot fully resolve dynamic quantization instability caused by cross-layer residual changes. Based on this analysis, we propose DynamicPTQ, a Dynamic Post-Training Quantization policy for phase-aware mixed-precision activation quantization. DynamicPTQ identifies quantization-sensitive layers from residual-stream dynamics and assigns 8-bit activation precision only to these layers, while keeping weights, KV caches, and other activations in 4-bit precision. It can be directly integrated with strong PTQ baselines such as QuaRot, SpinQuant, and FlatQuant. Experiments on LLaMA-2 and LLaMA-3 show that DynamicPTQ consistently improves perplexity and zero-shot QA performance under W4A4KV4 quantization, while achieving 1.05 to 1.07 times throughput improvement with modest memory overhead. These results demonstrate a practical path toward robust low-bit LLM inference.

Current virtual staining approaches offer the potential for time- and cost-efficient biomarker quantification in cancer diagnostics and prognostics. However, patch-wise inference for gigapixel whole slide images (WSIs) fails to maintain spatial continuity, yielding artifacts that cause catastrophic mismatches with ground-truth images. Although pathology Vision Foundation Models (VFMs) offer rich representations, their self-attention causes varying global contexts to produce inconsistent embeddings for the same physical region. We formalize and validate this ``context contamination'' as a sheaf-theoretic problem where these embeddings form a presheaf that violates the gluing axiom. To address this, we propose SheafStain, a new approach that reinterprets VFM features as sheaf-like sections for spatially and biologically coherent virtual staining. Specifically, SheafStain integrates class and patch tokens into a Schrödinger Bridge framework as sheaf-like sections. While the class token anchors biological consistency, patch tokens form a per-position spatial map. A backbone co-pretrained on Hematoxylin \& Eosin (H\&E) and Immunohistochemistry (IHC) yields non-degenerate cross-stain stalks, so a single VFM feature space supervises both input conditioning and output stain alignment. Departing from prior work that evaluates on isolated $256 \times 256$ patches and either random-crops or resizes the $1024 \times 1024$ ground truth, we translate at $256 \times 256$ and evaluate on the stitched $1024 \times 1024$ outputs across HER2, ER, PR, and Ki-67. SheafStain demonstrates promising results against six prior methods while mitigating patch-boundary stitching artifacts. Code will soon be released.

Condition-based Predictive Maintenance (PdM) for truck fleets has gained momentum in recent years. This maintenance strategy aims to minimize unplanned downtimes and reduce costs by monitoring the health status of vehicles and taking proactive action based on their condition. However, the implementation of condition-based PdM systems is challenging due to the large volume of data generated by the trucks, the inherent complexity of detecting failures through sensor data and the difficulties in finding cost-effective trade-offs in the solution's implementation. In this paper, we define and validate a condition-based PdM methodology built on the assumption that the wear-and-tear state of the monitored component can be represented as a monotonically non-decreasing time series. It involves selecting only the most recent observations from the time series and transforming them into a tabular format for classification using machine learning (ML) models designed for tabular data. Our results indicate that the proposed methodology reduces costs on the Scania Component X dataset compared to current state-of-the-art (SOTA) approaches, while also simplifying the modeling process through AutoML.

Continual anomaly detection in tabular data is challenging and remains largely underexplored, particularly in settings with heterogeneous feature schemas, distribution shifts, and severe class imbalance. In many real-world applications, data arrive sequentially from diverse domains, rendering conventional continual learning methods ineffective due to their reliance on a fixed input space. We propose a continual learning (CL) method, which can overcome these challenges and continually learn from different tasks. Our method consists of three main parts: our AGF model, Taskfusion augmentation, and outlier exposure. The AGF-model maps task-specific features into a shared space, then aligns distributions to reduce representation drift, and learns anomaly decision boundaries in the aligned space. To improve stability, we introduce Taskfusion augmentation, combining boundary-aware interpolation within tasks to refine the model anomaly boundaries and cross-task mixing to transfer anomaly structure across datasets. To handle class imbalance and memory constraints, we employ tabular dataset distillation to store compact synthetic replay samples, which are jointly used with augmented data in an outlier exposure objective for robust anomaly detection. We evaluate the approach on 21 heterogeneous datasets across multiple domains. Results show that our approach substantially improves continual anomaly detection performance over sequential fine-tuning and other CL baselines while reducing catastrophic forgetting and maintaining stable detection across heterogeneous datasets.

Low-light novel view synthesis is challenging because dark multi-view images contain noise, weak structural detail, and compressed dynamic range. Recent 3D Gaussian Splatting (3DGS) methods address these challenges by generating pseudo ground-truth (pseudo-GT) images as supervision targets when paired normal-light references are unavailable. Existing pseudo-GT methods apply a uniform linear gain to all pixels, which clips bright regions while providing insufficient enhancement in dark regions, limiting reconstruction quality. We observe that nonlinear tone mappings, long established in 2D low-light enhancement, have not been explored for pseudo-GT generation in 3D reconstruction. Accordingly, we propose a scene-adaptive nonlinear tone-curve framework that replaces linear pseudo-GT with nonlinear alternatives. The framework introduces percentile-based normalisation for scene-agnostic curve application, a scene-adaptive offset for automatic black-level adjustment, and two complementary curves: Adaptive SoftExp (ASE), a bounded exponential curve, and Adaptive Poly3 (AP3), a data-driven cubic polynomial. The module changes only the pseudo-GT computation and leaves the 3DGS backbone unchanged. Experiments on three benchmarks covering 21 scenes show that both curves consistently outperform the linear baseline with PSNR improvements up to +4.34 dB on LOM and +3.25 dB on RealX3D. Both curves achieve similar performance despite their different mathematical forms, suggesting the improvement is curve-agnostic. Code is available at https://github.com/lvmingzhe/adaptiveToneCurve

European Rail Traffic Management System (ERTMS) is a widely adopted standard unifying train management in the EU. While the standard allows for use cases like fully autonomous driving, cybersecurity has been an afterthought. Risk analysis enables the systematic assessment and prioritization of threats and mitigations. To date, it remains unclear which threats are most significant in ERTMS. This study systematically models components of ERTMS and analyzes their security in light of threats identified in the underlying technologies. The results suggest a concerning state of ERTMS, despite its critical role in railway safety. The use of legacy standards like EuroBalises and GSM-Railway (GSM-R) introduces vulnerabilities that persist across minimal ERTMS implementations, deployments incorporating various optional safety measures, and prospective future evolutions of the system, e.g., adopting Future Railway Mobile Communication System (FRMCS). Fully transitioning to European Train Control System (ETCS) level 2 was identified as the most significant measure for advancing ERTMS cybersecurity. The results indicate that a shift of ERTMS toward security is required to ensure availability and safe operation. While the chosen methodology proved its feasibility and shows remaining weaknesses of ERTMS, future work is needed to develop railway-centric adaptations to improve the quantification and evaluation of the computed risks.

Reward models for text-to-video (T2V) generation guide post-training but often fail at fine-grained semantic alignment. We trace this to two structural weaknesses in existing reasoning-based reward models: they do not systematically verify every condition described in the prompt, and the visual evidence supporting each judgment remains implicit in their free-form reasoning. We propose SG-PVR, a video reward model that addresses these limitations through plan-and-verify reasoning grounded in spatio-temporal scene graphs. The verification plan decomposes the prompt into atomic claims, ensuring every requirement is checked. The spatio-temporal scene graph, encoding entities, attributes, and temporally-grounded relations, is extracted from the video and maintained as a persistent structured visual reference throughout reasoning. Each claim is verified against both the video and the scene graph, anchoring judgments in explicit visual evidence. SG-PVR achieves strong performance on semantic alignment, including fine-grained temporal semantics. As a test-time reranker, it further enhances compositional alignment in T2V generation.

Open-vocabulary scene sketch semantic segmentation aims to assign dense semantic labels to sparse line drawings based on flexible category vocabularies specified at inference time, without relying on pixel-level annotations during training. Unlike natural images, sketches lack texture and color cues, making semantic understanding heavily dependent on stroke layout and spatial configuration, a challenge that renders single-layer vision-language features inherently unstable. Our key observation is that attention maps from different Vision Transformer layers encode complementary spatial cues: shallow layers capture global structural layouts, while deeper layers focus on local stroke intersections and object parts. This suggests that cross-layer aggregation provides a more robust structural prior than any individual layer alone. Leveraging this insight, we propose a structure-aware framework built upon \textbf{L}ayer-wise \textbf{A}ccumulated \textbf{S}tructural \textbf{A}ttention (\textbf{LASA}), which aggregates multi-layer attention to guide hierarchical semantic alignment under weak supervision and refine predictions during inference. Experiments on FS-COCO, SFSD, and FrISS show that LASA improves mIoU by $+3.43$, $+8.01$, and $+15.74$ over the prior weakly supervised baselines, demonstrating consistent gains in both segmentation accuracy and spatial coherence. Our source code will be made publicly available.

This paper presents a novel data-free and training-free compression approach for speech foundation models using channelwise clustering via k-means. More fine-grained, mixed sparsity pruning by layer-level varying number of parameter clusters is also explored. Experiments conducted on the LibriSpeech dataset suggest that when operating with pruning sparsity of 50% on HuBERT-large, consistent WER reductions of 27.73%/18.61% absolute (34.37%/21.91% relative) over the magnitude-based pruning were obtained on the test-clean and test-other subsets before fine-tuning and 0.19%/0.79% absolute (3.36%/4.62% relative) after fine-tuning with only 3 epochs. Similar WER reductions of 2.86%/5.02% absolute (59.21%/55.29% relative) were observed against magnitudebased pruning on Whisper-large-v3 at 10% sparsity, all with no significant WER increase relative to the uncompressed baseline.

Collecting participants' lived experiences is central to design research. Focus groups are uniquely valuable because participants not only share individual accounts but also respond to one another, surfacing comparison, disagreement, and collective sensemaking. However, focus groups are resource-intensive and highly sensitive to facilitation: moderators must probe for specificity, balance participation, manage topic flow, and sustain psychological safety, and subtle facilitation choices can shape what becomes salient. Recent HCI work and commercial meeting tools show that generative AI can scaffold live conversation through prompting, turn regulation, thematic mapping, and real-time summarization. Yet UXR teams lack a clear map of what these capabilities mean in focus groups and what methodological risks they introduce. We synthesize AI supports for live conversation and translate them into a focus-group-specific playbook organized by AI role (tool, co-host, host) and modality (text, voice, embodied).We synthesize prior work on AI-supported live conversation and propose a focus-group-specific playbook of AI supports organized by role (tool, co-host, host) and modality (text, voice, embodied). We characterize interactional trade-offs and identify open questions for evaluating AI-supported focus groups as methodological configurations.

Traceability information between requirements and source code greatly benefits the maintenance of a software system. Since manually establishing trace links is cumbersome and error-prone, previous research explored automated traceability link recovery (TLR) approaches to support this task. However, quality defects in requirements impact subsequent activities such as TLR, yet evidence about this remains scarce. Our objective is to contribute empirical evidence on this impact. At the same time, we aim to understand how the performance of TLR approaches varies given these quality defects. To this end, we annotated 28 types of quality defect in 189 use case descriptions from two datasets. Then, we executed five distinct TLR approaches on the dataset and measured their performance in recovering trace links. Finally, we performed statistical tests to quantify the defects' effect strength on this performance. Our results show that some quality defects harm TLR performance, e.g., sentences that do not start with noun phrases, while others actually benefit performance, e.g., use cases that include implementation details. Moreover, different types of approaches respond differently to these defects. As a consequence, the performance-optimizing choice of a TLR approach depends on the quality of the dataset.

Flow matching and diffusion models enable conditional generation across domains ranging from images to proteins, with recent extensions to out-of-distribution contexts. Yet generative models of neural time series have largely remained restricted to categorical conditioning, precluding compositional and zero-shot generalization. In this work, we propose a per-timestep conditioned diffusion transformer for generating realistic fMRI brain dynamics during unseen cognitive tasks by injecting both compositional language and optional spatial priors in-context. Such zero-shot generation could enable counterfactual neuroscience by supporting in-silico design and evaluation of novel cognitive experiments before empirical validation. Leveraging this model, we evaluate across hundreds of held-out task conditions and characterize predictive performance in relation to the training manifold. From language alone, the model recovers region-specific recruitment across tasks and held-out spatial activation patterns. Spatial priors, when available, complement the text pathway by anchoring generation in regions of task space where language alone degrades, while retaining the compositional structure needed for counterfactual task specification. To our knowledge this is the first generative model of whole-cortex fMRI dynamics for unseen cognitive tasks, advancing counterfactual neuroscience and data-driven experimental design.